Turning Movement Counts Research Articles

Real-Time Intersection Vehicle Turning Movement Counts from Live UAV Video Stream Using Multiple Object Tracking

Real-Time Intersection Vehicle Turning Movement Counts from Live UAV Video Stream Using Multiple Object Tracking

Real-Time Intersection Turning Movement Flow Forecasting Using a Parallel Bidirectional Long Short-Term Memory Neural Network Model

Real-time forecasting of intersection turning movements is a critical requirement for predictive traffic signal control at signalized intersections. Several traffic flow forecasting models have been developed in the past; however, most of them have focused only on the road segment-level traffic instead of the turning movement flow (TMF) at intersections. Therefore, in this paper, we propose a new TMF forecasting model, which consists of a combination of two neural network models, namely the stacked bidirectional long short-term memory and the traditional multi-layer perceptron model; this combination will enable effective learning for both short- and long-term time-varying patterns. Moreover, extensive computational experiments, using two years of turning movement counts at 22 intersections in the city of Milton, Ontario, Canada, explore the performance advantage of the proposed model in comparison with several state-of-the-art base models for forecasting accuracy, robustness, and transferability.

Intersection analysis using computer vision techniques with SUMO

Abstract This paper presents intersection analysis using computer vision techniques with Simulation of Urban MObility (SUMO). First, an efficient deep-visual tracking pipeline is proposed by using the off-the-shelf YOLO object detection architecture and cascading it with a discriminative correlation filter to produce reliable trajectories for traffic analysis of vehicles and pedestrians. While a variety of traffic measurements can be directly estimated from the extracted trajectories (e.g., speed, turning movement count), a method of incorporating turning movement count (TMC) within SUMO is proposed in order to mimic a realistic traffic flow for an observed intersection and its comprehensive analysis. Experimental evaluations on the developed tracking system implies that the YOLOv5 variant is the best for traffic cameras and, after appropriate fine-tuning using the University of Nevada, Las Vegas pedestrian data set, the YOLOv5 performance manifested a significant improvement with a recall value of 0.62. The tracking system is further employed for monitoring three other intersections in the downtown area of Las Vegas and turning movement counts were estimated for peak hours in the morning and evening of one day (7:00–9:00 and 16:00–18:00) at 15-min intervals. Finally, the intersection design, including traffic signals with estimated TMC, is used to calibrate SUMO to provide critical parameters (e.g., lane density, travel time, occupancy) for traffic signal performance evaluation and comprehensive intersection analysis. The signal design treatment demonstrates a significant improvement in travel times and simulation results indicate that the turning-left ratio is a crucial factor affecting the travel time of vehicles on each intersection leg.

Machine learning-based multi-target regression to effectively predict turning movements at signalized intersections

Effective prediction of turning movement counts at intersections through efficient and accurate methods is essential and needed for various applications. Commonly predictive methods require extensive data collection, calibration, and modeling efforts to estimate turning movements. In this study, three models were proposed to estimate turning movements at signalized intersections using approach volumes. Two sets of data from the United States and Canada were obtained to develop and test the proposed models. Machine learning-based regression models, including random forest regressor (RFR) and multioutput regressor (MOR) in addition to an artificial neural network (ANN) model, were developed and trained to analyze the relationship between approach volumes and corresponding turning movements. Multiple evaluation measurements were utilized to compare the models. All models produced satisfactory results. The RFR regression model outperformed the MOR model. However, the ANN model had the best performance when compared to the other models. The proposed models provide traffic engineers and planners with reliable and fast methods to estimate turning movements.

Development and Evaluation of Intersection-Based Turning Movement Counts Framework Using Two Channel LiDAR Sensors

Development and Evaluation of Intersection-Based Turning Movement Counts Framework Using Two Channel LiDAR Sensors

Network-level turning movement counts estimation using traffic controller event-based data

Accurate turning movement counts (TMCs) data collected from regional-wide signalized intersections is critical to regional transportation planning and simulation modeling. A variety of existing traffic sensors, configured at intersections for traffic detection and signal control, can generate a large amount of real-time high-resolution event-based data from traffic controllers but few of these sensors are configured to collect TMC. This paper proposes a methodology for estimating network-level TMC using existing traffic controller event-based data without installing additional sensors. First, relevant features that can indicate traffic arrival are extracted from existing event-based data, including detector occupancy time, detector-triggered count, and green time duration. With these features, a multi-output multilayer neural network model is developed to estimate TMC. To further improve network-level TMC estimation accuracy, intersection infrastructure data and point-of-interest (POI) data are included as exogenous variables for the proposed model. Ninety-three signalized intersections are chosen from the Pima County region, Arizona, to calibrate and verify the developed model. The validation results show that the proposed model can accurately estimate TMC, as indicated by a median Root Mean Square Error (RMSE) of 41 veh/15 min, 11 veh/15 min, and 12 veh/15 min for through movement, left-turn movement, and right-turn movement volume estimation, respectively. This research provides a new possibility of utilizing existing data sources to obtain network-level TMC data without additional infrastructure and labor costs for transportation agencies.

An Integrated Turning Movements Estimation to Petri Net Based Road Traffic Modeling

The tremendous increase in the urban population highlights the need for more efficient transport systems and techniques to alleviate the increasing number of the resulting traffic-associated problems. Modeling and predicting road traffic flow are a critical part of intelligent transport systems (ITSs). Therefore, their accuracy and efficiency have a direct impact on the overall functioning. In this scope, a new approach for predicting the road traffic flow is proposed that combines the Petri nets model with a dynamic estimation of intersection turning movement counts to ensure a more accurate assessment of its performance. Thus, this manuscript extends our work by introducing a new feature, namely turning movement counts, to attain a better prediction of road traffic flow. A simulation study is conducted to get a better understanding of how predictive models perform in the context of estimating turning movements.

Bicyclist Crash Comparison of Mixing Zone and Fully Split Phase Signal Treatments at Intersections with Protected Bicycle Lanes in New York City

This study builds on the success of New York City’s protected bicycle lanes (PBLs) through improving the understanding and impacts of different intersection designs. Specifically, to understand better the safety of the two primary intersection design treatments for single direction PBLs, the mixing zone and fully split phase signal, several different data sources are used in a before–after and cross-sectional analysis including; 184 intersections, nearly 17 years of reported bicycle injury crashes at intersections, and historic bicycle volume and turning movement counts. It was found that, following the installation of PBLs, the rate of crashes per bicyclist decreased by an average of 30%. The study also identified similar benefits between mixing zones and fully split phases at intersections with lower turn volumes while at higher turn volume locations the shorter mixing zones have the lowest crash rates. For wider cross-streets, the crash risk for bicycle riders increased regardless of the treatment, however fully split phase intersections perform better at wider crossings overall. Based on these results, it is recommended that the fully split phase treatment should be typically prioritized where the safety need is greatest, such as at wider intersections and two-way cross-streets. In addition, these findings reinforce the need to reduce turn speeds and cyclist exposure at multilane crossings.

Estimating Turning Movements at Signalized Intersections Using Artificial Neural Networks

Intersection turning movements’ counts are critical input data for traffic studies, analysis, and forecasting. These types of counts are often used to analyze operational performance of signalized intersections under different traffic conditions and peak hours. There are several methods to collect these movements, such as manual counts and video image processing. Traffic studies require turning movement counts although they may not be readily available, especially in the case of traffic forecasting or when seasonal adjustment factors are applied. In these cases, only approach volumes might be available and turning movements are assumed using different techniques to balance the inbound and outbound traffic. Nonetheless, these techniques require initial conditions, and their outcomes are highly sensitive to the initial assumptions. In this paper, a robust and practical method is developed. The turning movement counts at 691 four-leg, and 156 three-leg signalized intersections are analyzed to develop an accurate and reliable turning movement estimation model. A total of 4,175 hours of turning movement counts are used. An artificial neural networks (ANN) model is trained to analyze the relationship between the approach volumes and the corresponding turning movements. The results show that the developed ANN model can be utilized to predict turning movements at a high level of accuracy.

Use of Hi-resolution data for evaluating accuracy of traffic volume counts collected by microwave sensors

Abstract Over the past few years, the Utah Department of Transportation has developed the signal performance metrics (SPMs) system to evaluate the performance of signalized intersections dynamically. This system currently provides data summaries for several performance measures, one of them being turning movement counts collected by microwave sensors. As this system became public, there was a need to evaluate the accuracy of the data placed on the SPMs. A large-scale data collection was carried out to meet this need. Vehicles in the Hi-resolution data from microwave sensors were matched with the vehicles by ground-truth volume count data. Matching vehicles from the microwave sensor data and the ground-truth data manually collected required significant effort. A spreadsheet-based data analysis procedure was developed to carry out the task. A mixed model analysis of variance was used to analyze the effects of the factors considered on turning volume count accuracy. The analysis found that approach volume level and number of approach lanes would have significant effect on the accuracy of turning volume counts but the location of the sensors did not significantly affect the accuracy of turning volume counts. In addition, it was found that the location of lanes in relation to the sensor did not significantly affect the accuracy of lane-by-lane volume counts. This indicated that accuracy analysis could be performed by using total approach volumes without comparing specific turning counts, that is, left-turn, through and right-turn movements. In general, the accuracy of approach volume counts collected by microwave sensors were within the margin of error that traffic engineers could accept. The procedure taken to perform the analysis and a summary of accuracy of volume counts for the factor combinations considered are presented in this paper.

Optimized Video Tracking for Automated Vehicle Turning Movement Counts

This paper proposes a new method for automatically counting vehicle turning movements based on video tracking, expanding on previous work on optimization of parameters for road user trajectory extraction and on automated trajectory clustering. The counting method is composed of three main steps: an automated tracker that extracts vehicle trajectories from video data, an automated trajectory clustering algorithm, and an optimization algorithm. The proposed method was applied to obtain turning movement counts in three typical traffic engineering case studies in Canada representing industry-type conditions. These exhibited varying levels of tracking difficulty, ranging from a single-lane off-ramp to a six-movement intersection with a stop and a right-turn channel. Because of a limitation of the data set, giving flows per movement and not per lane, all sites were chosen with a single lane per movement. The 3-h morning peak period was used in the case studies. The results show an average weighted generalization error of 12% for more than 3,700 vehicles automatically analyzed for more than 8 h of video, ranging from 9.5% to 19.5%. The generalization error is on average 8.6% (and as low as 6.0% per movement) for the 3,084 uninterrupted vehicles that are in plain view of the camera. This paper describes in detail the methodology used and discusses the factors that affect counting performance and how to improve counting accuracy in further research.

Assessment of the Robustness of Signal Timing Plans in an Arterial Corridor Through Seasonal Variation of Traffic Flows

Arterial traffic signal systems, predominantly in the United States, deploy multiple signal timing plans to account for daily variability of traffic demand. Those types of traffic flow deviations should be anticipated when timing plans are designed and, therefore, serviced satisfactorily. When traffic flow patterns are no longer predictable, a predetermined time-of-day (TOD) plan may no longer be the optimal one. This research aimed to examine signal timing optimality by applying a method similar to the selection of a traffic responsive plan to recognize automatically the best timing plan suited to current traffic conditions. The proposed method attempted to determine whether the optimality of signal timing settings could have been effectively estimated when systematic detector counts of the major approach were available. The study used 4 months of data from field microwave detectors coupled with data of turning-movement counts obtained over several days. The findings show that TOD signal timing plans mainly depended on adequate data collection that best describes a specific set of traffic conditions. Thus, the designed plan was as optimal as the related traffic information was reliable, whereas a problem arose in the case of limited-availability and low-quality data. New technologies are capable of collecting and storing massive amounts of data. Even if the granularity of collected data is low, the data can be used to improve traffic performance (i.e., reduce corridor delay). This realization could be of particular importance to traffic agencies that have installed, or plan to install, new field devices.

Innovative detector layout for automated traffic turning volume counting

SummaryBecause of many advantages, loop detectors are the most common practice for obtaining data to control intersections. However, they have some drawbacks, including the fact that multiple detectors are usually required to monitor a location. The current practice in many cities is to install four consecutive loop detectors per lane, or two at the stop bar and one as an advanced detector. In some cities, there are also departure detectors. All these configurations have some practical problems and do not produce accurate counts especially in shared lanes. In this paper, a new placement configuration for departure detectors is proposed and named the mid‐intersection detector (MID). In this configuration, departure detectors are moved back to the middle of the intersection in such a way that they can be activated by more than one movement at different times. In some cases, departure detectors lack equations for calculating turning movements, a problem solved by MIDs because each movement passes more detectors along its path (without increasing the number of loops), and therefore they can produce more accurate and reliable data for obtaining turning movement counts. Copyright © 2017 John Wiley & Sons, Ltd.

Using stop bar detector information to determine turning movement proportions in shared lanes

SummaryTurning vehicle volumes at signalized intersections are critical inputs for various transportation studies such as level of service, signal timing, and traffic safety analysis. There are various types of detectors installed at signalized intersections for control and operation. These detectors have the potential of producing volume estimates. However, it is quite a challenge to use such detectors for conducting turning movement counts in shared lanes. The purpose of this paper was to provide three methods to estimate turning movement proportions in shared lanes. These methods are characterized as flow characteristics (FC), volume and queue (VQ) length, and network equilibrium (NE). FC and VQ methods are based on the geometry of an intersection and behavior of drivers. The NE method does not depend on these factors and is purely based on detector counts from the study intersection and the downstream intersection. These methods were tested using regression and genetic programming (GP). It was found that the hourly average error ranged between 4 and 27% using linear regression and 1 to 15% using GP. A general conclusion was that the proposed methods have the potential of being applied to locations where appropriate detectors are installed for obtaining the required data. Copyright © 2016 John Wiley & Sons, Ltd.

Automated Turning Movement Counts for Shared Lanes: Leveraging Vehicle Detection Data

Turning movement count data (i.e., vehicle volumes broken down by movement, approach, and time periods) are the foundation of signal performance evaluations and a crucial component of a data-driven decision-making process used by transportation agencies. In this paper, the authors show how data available from intersections equipped with radar-based vehicle detection can be used to produce turning movement counts. A classification algorithm developed and discussed by the authors is capable of producing turning movement counts regardless of lane configuration and without the need for definition of detection zones. The algorithm works by using the data produced by vehicle detection systems that go unused and are never communicated to the signal controller. The nature of the data collection process makes the algorithm independent of the controller used. Results from the algorithm are promising; an average error of −0.26 vehicles per 15-min count period (absolute error of 2.31 vehicles) was obtained with the algorithm. Furthermore, the application of the algorithm provides an opportunity beyond signal operations. Processed trajectory data and results from the algorithm could be used to break the boundary that often exists between operations, planning, and safety, and thus show how a monitoring system that relies on the algorithm could help a traffic monitoring program meet the different—and sometimes competing—interests of agencies.

Vision-Based Turning Movement Monitoring:Count, Speed & Waiting Time Estimation

Abstract-This work presents a vision-based vehicle tracking system designed for improved intersection behavior analysis. Motion-based tracking is paired with trajectory analysis methods for accurate turning movement counts of all legs of an intersection using a single camera. Additional behavior analysis methods are implemented to estimate both speed and waiting time at intersections. Improved count accuracy is obtained through cooperation between complementary counting modules and a path reconstruction module is integrated to reconnect broken tracks of the same stationary vehicles at intersections. Experimental results show around 15% improvement in counting accuracy and different behavior of right turns were observed by speed profile and waiting time estimation. Right turns are taken more slowly than other turns and vehicle waiting times are shorter at signalized intersections since they do not always need permission from the signal. We finally show that the waiting time of right turns are directly affected by opposing flows regardless of signal phase.

Real-time estimation of turning movement counts at signalized intersections using signal phase information

A variety of sensor technologies, such as loop detectors, traffic cameras, and radar have been developed for real-time traffic monitoring at intersections most of which are limited to providing link traffic information with few being capable of detecting turning movements. Accurate real-time information on turning movement counts at signalized intersections is a critical requirement for applications such as adaptive traffic signal control. Several attempts have been made in the past to develop algorithms for inferring turning movements at intersections from entry and exit counts; however, the estimation quality of these algorithms varies considerably. This paper introduces a method to improve accuracy and robustness of turning movement estimation at signalized intersections. The new algorithm makes use of signal phase status to minimize the underlying estimation ambiguity. A case study was conducted based on turning movement data obtained from a four-leg signalized intersection to evaluate the performance of the proposed method and compare it with two other existing well-known estimation methods. The results show that the algorithm is accurate, robust and fairly straightforward for real world implementation.

A generalized mathematical model to determine the turning movement counts at roundabouts

Traffic turning movement counts at roundabouts is one of the key inputs required for roundabout assessment, control and management. Traditionally, a direct counting is conducted to track a vehicle from entering through circulation until exiting. This counting may be difficult and costly due to the size of roundabout, the vision obstacles, and the continuous traffic flow. Many researchers tried to avoid the tracking problem by counting only at entries and exits, then estimating the movements based on historical data which unfortunately affect the results. Other researchers reduced the tracking problem by counting some turning movements in addition to at entries and exits, then calculating mathematically the remaining movements. This approach is practical and accurate; however, it was applied on limited cases. In this paper, a generalized mathematical model was developed to calculate the most difficult movements based on the easiest movements determined based on the size of monitoring area. The developed model can be used to calculate the turning movements, including the u-turns, for roundabouts with any number of legs. The developed model was presented in O–D matrix forms to be practical and user-friendly. The model was validated against reference count data and the results were found to be satisfactory.

The influence of neighborhood traffic density on the respiratory health of elementary schoolchildren

Background Several studies have found that living near major roadways is associated with an increase in respiratory illness but few studies have measured the volume and type of traffic. Objective We investigated the relation between traffic volume and respiratory health of 2328 children 9 to 11 years old in the city of Windsor, Canada. Methods We identified the roadways within a 200 meter radius of the child's neighborhood using the latitude and longitude of the residential postal code. Traffic exposure was defined as the sum of the annual volume of vehicles on all of these roadways. Volume was calculated using sensors to detect passing vehicles (simple traffic counts), and by counts and direction of traffic at intersections (turning movement counts). Ventilatory lung function was measured by spirometry and airway inflammation by exhaled nitric oxide (eNO). Results The odds ratio between an interquartile increase in truck turning movement counts and chest congestion was 1.20 (1.06–1.35). The percentage of predicted FVC declined 0.68%, (95% CI 1.32, 0.03) for an interquartile increase in simple traffic counts (33,787 vehicles daily). Among those with self-reported asthma, effect sizes were larger. Percentage predicted FEV 1 declined 1.84% (95% CI 0.07, 3.61) associated with an interquartile range increase in turning movement counts. No statistically significant change was detected between traffic measures and exhaled nitric oxide. Conclusions Our findings provide further support for the hypothesis that neighborhood exposure to traffic-related air pollution increases respiratory symptoms and reduces ventilatory function in children, especially those with self-reported asthma.

Data Collection Methodology for Container Truck Traffic in Inland Port Cities

This paper describes a data collection methodology to address insufficient data sources for estimation of urban container truck traffic (drayage) volumes. The methodology is sensitive to the characteristics of drayage and offers a systematic approach for acquiring container truck traffic data for constructing models to estimate drayage volumes. The methodology consists of (a) acquiring urban truck traffic estimates and national and provincial- or state-level container traffic databases, (b) characterizing shippers and carriers through field investigations and surveys, and (c) designing a container truck data collection program. Short-term manual truck classification intersection turning movement counts were conducted to obtain body style and axle configuration data for articulated trucks. Temporal expansion factors were developed and applied to short-term count data to produce average daily container truck traffic volume estimates and reveal temporal, physical, and spatial distribution differences between container trucks and other articulated trucks. The paper provides a rationale for selecting count station locations and their temporal characteristics, choosing the number of counts and their duration, determining the types of data to collect, and identifying container generators. The methodology is generally applicable to North American inland port cities. The data feed a model that is intended to assist transportation engineers in understanding urban drayage operations and quantifying the exposure of these trucks.